| { | |
| "best_metric": 0.8304761904761905, | |
| "best_model_checkpoint": "final_models/glue_models/structroberta_s2_final//finetune/sst2/checkpoint-2000", | |
| "epoch": 9.47867298578199, | |
| "global_step": 4000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.95, | |
| "eval_accuracy": 0.8110235929489136, | |
| "eval_f1": 0.8102766798418972, | |
| "eval_loss": 0.4257064461708069, | |
| "eval_mcc": 0.6220413857242502, | |
| "eval_runtime": 0.9922, | |
| "eval_samples_per_second": 511.972, | |
| "eval_steps_per_second": 64.5, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "learning_rate": 4.407582938388626e-05, | |
| "loss": 0.3763, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "eval_accuracy": 0.8070865869522095, | |
| "eval_f1": 0.8243727598566307, | |
| "eval_loss": 0.5127468705177307, | |
| "eval_mcc": 0.6277527819251606, | |
| "eval_runtime": 0.9929, | |
| "eval_samples_per_second": 511.623, | |
| "eval_steps_per_second": 64.456, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.37, | |
| "learning_rate": 3.815165876777251e-05, | |
| "loss": 0.1968, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "eval_accuracy": 0.8129921555519104, | |
| "eval_f1": 0.8080808080808081, | |
| "eval_loss": 0.5491798520088196, | |
| "eval_mcc": 0.6265026705560554, | |
| "eval_runtime": 0.9941, | |
| "eval_samples_per_second": 511.013, | |
| "eval_steps_per_second": 64.38, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 3.55, | |
| "learning_rate": 3.222748815165877e-05, | |
| "loss": 0.1208, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 3.79, | |
| "eval_accuracy": 0.8188976645469666, | |
| "eval_f1": 0.8270676691729323, | |
| "eval_loss": 0.6646919846534729, | |
| "eval_mcc": 0.6413015929266274, | |
| "eval_runtime": 0.9945, | |
| "eval_samples_per_second": 510.824, | |
| "eval_steps_per_second": 64.356, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 4.74, | |
| "learning_rate": 2.6303317535545023e-05, | |
| "loss": 0.0765, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 4.74, | |
| "eval_accuracy": 0.8248031735420227, | |
| "eval_f1": 0.8304761904761905, | |
| "eval_loss": 0.687165379524231, | |
| "eval_mcc": 0.6515283866658971, | |
| "eval_runtime": 0.9913, | |
| "eval_samples_per_second": 512.447, | |
| "eval_steps_per_second": 64.56, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 5.69, | |
| "eval_accuracy": 0.8031495809555054, | |
| "eval_f1": 0.7975708502024291, | |
| "eval_loss": 0.790716826915741, | |
| "eval_mcc": 0.6068978167568908, | |
| "eval_runtime": 0.9943, | |
| "eval_samples_per_second": 510.929, | |
| "eval_steps_per_second": 64.369, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 5.92, | |
| "learning_rate": 2.037914691943128e-05, | |
| "loss": 0.0477, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 6.64, | |
| "eval_accuracy": 0.8248031735420227, | |
| "eval_f1": 0.8271844660194174, | |
| "eval_loss": 0.8302326202392578, | |
| "eval_mcc": 0.6500578442769696, | |
| "eval_runtime": 0.991, | |
| "eval_samples_per_second": 512.589, | |
| "eval_steps_per_second": 64.578, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 7.11, | |
| "learning_rate": 1.4454976303317535e-05, | |
| "loss": 0.0294, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 7.58, | |
| "eval_accuracy": 0.8129921555519104, | |
| "eval_f1": 0.8118811881188119, | |
| "eval_loss": 0.9771924614906311, | |
| "eval_mcc": 0.6259775087526785, | |
| "eval_runtime": 0.9936, | |
| "eval_samples_per_second": 511.275, | |
| "eval_steps_per_second": 64.413, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 8.29, | |
| "learning_rate": 8.530805687203793e-06, | |
| "loss": 0.0172, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 8.53, | |
| "eval_accuracy": 0.8188976645469666, | |
| "eval_f1": 0.8210116731517509, | |
| "eval_loss": 1.0711852312088013, | |
| "eval_mcc": 0.6381511040271763, | |
| "eval_runtime": 0.9924, | |
| "eval_samples_per_second": 511.908, | |
| "eval_steps_per_second": 64.492, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 9.48, | |
| "learning_rate": 2.6066350710900472e-06, | |
| "loss": 0.0113, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 9.48, | |
| "eval_accuracy": 0.8149606585502625, | |
| "eval_f1": 0.815686274509804, | |
| "eval_loss": 1.136107087135315, | |
| "eval_mcc": 0.6300165875296968, | |
| "eval_runtime": 0.9923, | |
| "eval_samples_per_second": 511.934, | |
| "eval_steps_per_second": 64.496, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 9.48, | |
| "step": 4000, | |
| "total_flos": 3.997142324210074e+16, | |
| "train_loss": 0.10950899147987365, | |
| "train_runtime": 2086.3883, | |
| "train_samples_per_second": 242.179, | |
| "train_steps_per_second": 2.023 | |
| } | |
| ], | |
| "max_steps": 4220, | |
| "num_train_epochs": 10, | |
| "total_flos": 3.997142324210074e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |